STRTSSkin Temperature Controlled Repeated Thermal Stimulation System
Copyright 1988-2018, All rights reserved.
References in periodicals archive ?
Comparatively, STRTs were larger in size, more complex in nature, and less interpretable because they captured more detailed information about the response variables.
The predictive performance of both MTRTs and STRTs for all infection levels is relatively strong.
In this study, RF-MTRTs perform better than both STRTs and MTRTs for all infestation levels (Table 2; Table 3).
MTRTs that predicted multiple biocontrol measures simultaneously were smaller and more interpretable than STRTs for all infection levels.
Caption: Figure 3: The most influential bands used to construct (a) STRTs to predict RLCC and (b) STRTs to predict FSD for the low, medium, and high infestation levels.
Caption: Figure 4: Individual STRTs used to predict (a) FSD and (b) RLCC from canopy spectral reflectance for the high infestation level.
Table 2: Predictive performance of the STRTs for the low, medium, and high infestation levels.
One such technique is single target regression trees (STRT) conducting binary recursive partitioning producing a set of rules and a regression model to predict a single response variable [27, 28].
Model interpretability has been evaluated by determining the size of STRT and MTRT after pruning.
[30] and used in other studies [29, 34, 35], the predictive performance of the STRT, MTRT, and RF-MTRT was evaluated by computing the Pearson correlation coefficient and root mean square error (RMSE).
The size of STRT models was used as a measure of interpretability because the deeper the model, the more numerous and more complex the decision rules are, thereby decreasing the interpretability of the model [29].
Even though STRT models were difficult to interpret, each STRT model was inspected to determine key spectral regions and identify influential bands that were used as decision rules to construct the models.